Open Access
Available online http://arthritis-research.com/content/11/1/R1
Page 1 of 10
(page number not for citation purposes)
Vol 11 No 1
Research article
Interactions among type I and type II interferon, tumor necrosis
factor, and -estradiol in the regulation of immune
response-related gene expressions in systemic lupus
erythematosus
Hooi-Ming Lee
1
, Toru Mima
1
, Hidehiko Sugino
1
, Chieko Aoki
1
, Yasuo Adachi
1
, Naoko Yoshio-
Hoshino
1
, Kenichi Matsubara
2
and Norihiro Nishimoto
1
1
Laboratory of Immune Regulation, Graduate School of Frontier Biosciences, Osaka University, 1-3 Yamada-Oka, Suita City, Osaka 565-0871, Japan
2
DNA Chip Research Incorporated, 1-1-43 Suehirocho, Tsurumi-ku, Yokohama, Kanagawa 230-0045, Japan

these genes, several networks including cytokines – such as
TNF and IFN – and E2 were constructed. TNF-regulated genes
were dominant in these networks, but in vitro TNF stimulation on
peripheral blood mononuclear cells showed no differences in
the above gene expressions between SLE and healthy
individuals. Co-stimulating with IFN and one of TNF, IFN, or
E2 revealed that TNF has repressive effects while IFN
essentially has synergistic effects on IFI gene expressions in
vitro. E2 showed variable effects on IFI gene expressions among
three individuals.
Conclusions TNF may repress the abnormal regulation by IFN
in SLE while IFN may have a synergistic effect. Interactions
between IFN and one of TNF, IFN, or E2 appear to be involved
in the pathogenesis of SLE.
Introduction
Systemic lupus erythematosus (SLE) is a prototypical autoim-
mune disease characterized by multiple organ damage, high
titers of autoantibodies, and various clinical manifestations [1].
Numerous disorders in the immune system and abnormalities
in cytokine productions have been described in patients with
SLE. The exact pathological mechanisms are still obscure,
however, and the roles of the cytokines are not well under-
stood. High levels of TNF, type I interferon, and type II inter-
feron in the sera of patients with SLE have been reported [2-
4]. On the other hand, an impaired production of IL-12 by T
lymphocytes from SLE patients in vitro has also been
aRNA: amino allyl RNA; Ct: cycle threshold; E2: -estradiol; FcR: Fc receptor; GBP: guanylate binding protein; HLA: human leukocyte antigen; IFI:
interferon-inducible; IFIT: interferon-induced protein with tetratricopeptide repeats; IFN: interferon; IL: interleukin; IRF7: interferon regulatory factor 7;
ISG15: interferon-stimulated gene, 15 kDa; MAPK: mitogen-activated protein kinase; MHC: major histocompatibility complex; NFB: nuclear factor
of kappa light polypeptide; OAS1: 2',5'-oligoadenylate synthetase 1; OASL: 2',5'-oligoadenylate synthetase-like; PBMC: peripheral blood mononu-

municating with each other, we performed a network-based
analysis to identify aberrant regulations or interactions among
differentially expressed molecules observed in this study. We
also investigated the effect of interactions between IFN and
one of TNF, IFN, or -estradiol (E2) on the expression of these
molecules.
Materials and methods
Patients and healthy individuals
Eleven patients (all women, median age 35 years, range 27 to
72 years) with SLE fulfilled by the diagnostic criteria of the
American College of Rheumatology [14] and six healthy
women were enrolled in the present study after obtaining their
written informed consent. The study was approved by the Eth-
ical Committee of Osaka University Medical School for clinical
studies on human subjects.
The majority of the SLE patients (n = 10) were treated with
<20 mg/day prednisolone. Three of these 10 patients were
treated with one of cyclosporine, azathioprine, or methotrexate
in combination with prednisolone, respectively. The remaining
patient was treated with >20 mg/day prednisolone.
The median disease activity score of SLE patients based on
the SLE Disease Activity Index 2000 instrument [15] was 10
(range 6 to 24). Two patients had very active states (SLE Dis-
ease Activity Index 2000 score >12) while the other patients
had active states (SLE Disease Activity Index 2000 score = 4
to 12). The median of the assessment based on the BILAG
index [16] was 4 (range 1 to 13).
Meanwhile, the median of the total white blood cells for the
patients was 6,160 (range 4,840 to 12,230). The median of
the total number (proportion) of neutrophils was 4,919

group of samples were selected for further analysis. The
microarray data have been deposited in NCBIs Gene Expres-
sion Omnibus [GEO:GSE12374].
Gene ontology and network-based analysis
Genes identified to be differentially expressed by >10%
according to the microarray analysis with a median signal
intensity difference of at least 100 between the SLE patient
and healthy individual groups (in order to reduce errors per-
taining to low-level expression at close to noise level) were
functionally categorized using Expression Analysis Systematic
Explorer version 2.0 bioinformatics software [17,18]. Interac-
tions among the differentially expressed genes in the func-
tional category of immune response were investigated through
the use of Ingenuity Pathway Analysis version 5.5 [19]. Net-
works generated by less than five uploaded genes were
excluded from the analysis.
Available online http://arthritis-research.com/content/11/1/R1
Page 3 of 10
(page number not for citation purposes)
Stimulation of peripheral blood mononuclear cells
To assess TNF signaling, PBMCs from six patients diagnosed
with SLE and from three healthy individuals were utilized. All
PBMCs used in the experiments were isolated from
heparinized whole blood using a Ficoll-Paque™ Plus (GE
Healthcare Biosciences, Uppsala, Sweden) gradient centrifu-
gation according to the manufacturer's recommendations. The
cells were incubated in RPMI 1640 with 10% heat-inactivated
fetal bovine serum and TNF (20 ng/ml) in a carbon dioxide
incubator at 37°C for 24 hours.
To examine the effects of interactions between IFN and one

cycle), 10 minutes at 95°C (one cycle), 15 seconds at 95°C
and 1 minute at 60°C (40 cycles). For each gene (performed
in duplicate for each sample), cycle threshold (Ct) values were
determined from the linear region of the amplification plot and
were normalized by subtracting the Ct value of GAPDH (gen-
erating a Ct value). The response to the cytokines or E2 was
determined by subtracting the Ct value for the time-matched
control from the Ct value for the stimulated sample (Ct
value). The fold change was subsequently calculated using the
formula 2Ct (where Ct was converted to an absolute
value).
Statistical analysis
The unpaired Mann-Whitney test was used to determine sta-
tistically significant differences in the mRNA expression levels
between the SLE patient and healthy individual groups. The
criterion for the statistical significance was P < 0.05.
Results
Immune response-related genes identified by gene
ontology analysis
Thirty-eight downregulated genes and 68 upregulated genes
were categorized into the functional category of immune
response. Most of the 68 upregulated genes were interferon
regulated – including 17 IFI genes such as interferon-induced
protein with tetratricopeptide repeats (IFIT) 1, 2',5'-oligoade-
nylate synthetase 1 (OAS1), 2',5'-oligoadenylate synthetase-
like (OASL), interferon-stimulated gene, 15 kDa (ISG15), and
interferon regulatory factor 7 (IRF7) that have been reported
as overexpressed in the PBMCs of SLE.
Network-based analysis on the downregulated or
upregulated genes in the functional category of immune

TLR2, TLR4, and TLR5) and another cluster of Fc receptors
(FcRs) were identified in this network. The two clusters were
indirectly connected through p38 MAPK and NFB, suggest-
ing there may be functional interactions among these mole-
cules through this pathway. This network was overlapped with
Arthritis Research & Therapy Vol 11 No 1 Lee et al.
Page 4 of 10
(page number not for citation purposes)
the fourth network, whose central molecules were IFN and E2
(Figure 2d). There were nine IFI molecules found in the first
and fourth networks. The second network was represented by
Akt and a calcium ion at the center (Figure 2b), while the third
network was mainly attributed to TNF (Figure 2c). We found
that two IFI molecules were included in the second network,
and that seven out of the 14 upregulated molecules that con-
structed the third network were IFI molecules.
Gathering the above results, TNF, IFN, and E2 were depicted
by both downregulated and upregulated molecules in the net-
works. As most of the genes in the immune response were
TNF regulated, we performed stimulating experiments on the
PBMCs of SLE patients and healthy individuals to assess the
TNF regulation on the immune response-related molecules in
SLE. On the other hand, although the expression of IFN was
not upregulated and was not depicted in networks related to
TNF, IFN, or E2, IFI molecules were found ranging over the
four networks. Furthermore, it has been reported that there
exist elevated levels of type I interferon in the SLE serum. Type
I interferon therefore appears to have complicated interactions
with various cytokines and E2. This encouraged us to further
examine the effects of interactions between IFN and one of

2',5'-oligoadenylate synthetase-like (OASL) Hs00388714_m1
Guanylate binding protein 1 (GBP1) Hs00266717_m1
Guanylate binding protein 2 (GBP2) Hs00269759_m1
IL8RA Hs00174146_m1
C-type lectin domain family 4, member E (CLEC4E) Hs00372017_m1
TNF-induced protein 6 (TNFAIP6) Hs00200180_m1
Available online http://arthritis-research.com/content/11/1/R1
Page 5 of 10
(page number not for citation purposes)
(GBP) 1, GBP2, IL8RA, C-type lectin domain family 4 member
E, and TNF-induced protein 6), all of which were TNF regu-
lated, were selected and their mRNA expressions upon TNF
stimulation were measured by quantitative RT-PCR. All of the
genes selected showed essentially the same responses to
TNF stimulation on PBMCs independent of the individual (Fig-
ure 3). CD40, IL12B, prostaglandin E synthase, C-type lectin
domain family 4 member E, and TNF-induced protein 6 were
upregulated, while CD1C, IFIT1, IFIT3, OAS1, and IL8RA
were downregulated upon TNF stimulation in both SLE
patients and healthy individuals.
The in vivo gene expression profiles of SLE, however, were dif-
ferent from the results of in vitro PBMC stimulation by TNF. For
example, CD40 was downregulated in vivo but was upregu-
lated upon TNF stimulation in vitro. Meanwhile, IFI genes such
as IFIT1, IFIT3, OAS1, ISG15 and IRF7, and IL8RA were
upregulated in vivo, but IFIT1, IFIT3, OAS1 and IL8RA were
downregulated, while ISG15 and IRF7 showed almost no
response to TNF in vitro. These data suggest that other solu-
ble factors might be involved in the regulation on the gene
expression. Indeed, high levels of interferon in SLE serum have

The relative expressions of three of the representative genes
(that is, IFIT1, IFIT3, and IFI27) from three women are shown
in Figure 4. A remarkable suppression was observed through
the TNF and IFN co-stimulating experiment (Figure 4a). On
the other hand, there was synergism between IFN and IFN
on IFI gene expressions, although with some exceptions like
IFIT1 (Figure 4b). IFIT1 was downregulated upon IFN and
IFN co-stimulation, unlike stimulation with IFN alone. E2
showed no significant or consistent interaction with IFN for
most of the IFI genes. Inconsistent responses to E2 stimula-
tion, however, were observed among the three healthy donors
on IFI27. E2 tended to downregulate IFI27 expression in one
donor but upregulated expression in the other two donors (Fig-
ure 4c).
To test a hypothesis that TNF decreases IFI gene expression
through suppressing IFN production, we examined the effect
of TNF or IFN on IFN mRNA expression. Its expression was
too low to be measured and there were no significant changes
in TNF, IFN, or TNF + IFN 24-hour-stimulated PBMCs.
Figure 2
Network-based analysis of upregulated genes in the functional category of immune responseNetwork-based analysis of upregulated genes in the functional category of immune response. (a) Network 1, (b) Network 2, (c) Network 3,
and (d) Network 4 constructed by upregulated genes.
Available online http://arthritis-research.com/content/11/1/R1
Page 7 of 10
(page number not for citation purposes)
Discussion
To identify the molecules involved in the aberrant immune sys-
tem of SLE, we compared the gene expression profiles of
peripheral blood between SLE patients and healthy individuals
using microarray technology followed by gene ontology analy-

gested the possibility that TLR7 may activate cells through
similar pathways [25]. Although in our study overexpression of
DNA-recognizing TLR9, which has been suggested to be trig-
gered by immune complexes containing DNA in SLE [26,27],
was not statistically significant according to the rank test,
seven out of the 11 SLE patients showed upregulated expres-
sions of TLR9. In addition, TLR1, TLR2, TLR4, and TLR5 –
which serve to recognize bacterial components such as
lipopolysaccharide or lipopeptides [28,29] – were also upreg-
ulated. Our network-based analysis therefore suggested the
hypothesis that the interaction between TLRs and FcRs is
involved in the pathogenesis of SLE.
We additionally found that networks whose central molecule
was TNF, IFN, or E2 were represented by both the downreg-
ulated genes and the upregulated genes in the functional cat-
egory of immune response. This observation suggested that
TNF, IFN, or E2 may be involved in the abnormal expressions
of both downregulated and upregulated genes in the immune
response. Indeed, the elevated level of some cytokines such
as TNF and interferon in the sera of SLE patients has been
reported [2,4,30,31]. Although our data did not show a signif-
icant increase in the gene expressions of TNF, IFN, or IFN in
themselves according to rank test, more than one-half of the
SLE patients' individual data showed an increase in the TNF
gene expression in our study (data not shown). For IFN, the
expression was not increased in the peripheral blood but it
may be produced at the other site. Siegal and colleagues have
demonstrated that purified interferon-producing cells were
CD4
+

poorly understood, we suspect that the elevated expression of
TNF in SLE reduces the overexpression of IFI genes. Since
serum levels of both TNF and IFN were reportedly elevated
in SLE, as mentioned above, it is possible that the increased
serum TNF level in SLE is an outcome to compensate the
immune system balance altered by IFN in SLE. Consider that
patients with rheumatoid arthritis or Crohn's disease under
TNF-blocking therapies can develop autoantibodies to nuclear
antigens [34]; therapeutic TNF blockades could thus lead to
an exacerbation of certain autoimmune diseases such as SLE
and to provoke lupus-like manifestations. Palucka and col-
leagues reported recently that blocking TNF signaling
increases the production of IFN by plasmacytoid dendritic
cells and induces an IFN signature in the blood of arthritis
patients [35]. This may be another mechanism for TNF inhibi-
tor to induce the IFN signature. We confirmed that there was
no significant effect, however, of TNF on IFN gene expres-
sion in the PBMCs in our experiment. Furthermore, the 500
units/ml IFN we used for stimulation is obviously a higher
amount than endogenously produced IFN. TNF therefore
appeared to directly suppress IFI gene expression in PBMCs.
We suggest that the direct suppressive effect of TNF on the
IFN signature induced by IFN, at least, exists in the network
regulation of cytokines in vivo.
The results of the co-stimulating experiments did not show any
strong evidence of a functional interaction between E2 and
IFN on the expression of IFI genes. Inconsistent gene expres-
sion patterns were observed in the co-stimulating experiments,
possibly due to the hormonal effects of the women donors.
The modulation of estrogens on humoral immune response

assisted with data analysis. CA performed labeling and scan-
ning of the microarrays. YA assisted with data analysis. NY-H
assisted with data analysis. KM assisted in microarray data
acquirement. NN designed the study, enrolled patients, and
assisted with data analysis and interpretation. All authors read
and approved the final manuscript.
Acknowledgements
The present work was supported by grants from the Ministry of Health,
Labor and Welfare of Japan. The authors would like to thank Ms Tami
Nanga for excellent secretarial support.
References
1. Kotzin BL: Systemic lupus erythematosus. Cell 1996,
85:303-306.
2. Gabay C, Cakir N, Moral F, Roux-Lombard P, Meyer O, Dayer JM,
Vischer T, Yazici H, Guerne PA: Circulating levels of tumor
necrosis factor soluble receptors in systemic lupus erythema-
tosus are significantly higher than in other rheumatic diseases
and correlate with disease activity. J Rheumatol 1997,
24:303-308.
3. Kim T, Kanayama Y, Negoro N, Okamura M, Takeda T, Inoue T:
Serum levels of interferons in patients with systemic lupus
erythematosus. Clin Exp Immunol 1987, 70:562-569.
4. Hooks JJ, Moutsopoulos HM, Notkins AL: Circulating interferon
in human autoimmune diseases. Tex Rep Biol Med 1981,
41:164-168.
5. Horwitz DA, Gray JD, Behrendsen SC, Kubin M, Rengaraju M, Oht-
suka K, Trinchieri G: Decreased production of interleukin-12
and other Th1-type cytokines in patients with recent-onset
systemic lupus erythematosus. Arthritis Rheum 1998,
41:838-844.

ada Y, Deguchi H, Suemura M, Miyake T, Kawase I, Zhao H, Tomi-
yama Y, Saeki Y, Nojima H: Isolation and expression profiling of
genes upregulated in the peripheral blood cells of systemic
lupus erythematosus patients. DNA Res 2005, 12:429-439.
13. Feng X, Wu H, Grossman JM, Hanvivadhanakul P, FitzGerald JD,
Park GS, Dong X, Chen W, Kim MH, Weng HH, Furst DE, Gorn A,
Mc Mahon M, Taylor M, Brahn E, Hahn BH, Tsao BP: Association
of increased interferon-inducible gene expression with dis-
ease activity and lupus nephritis in patients with systemic
lupus erythematosus. Arthritis Rheum 2006, 54:2951-2962.
14. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF,
Schaller JG, Talal N, Winchester RJ: The 1982 revised criteria for
the classification of systemic lupus erythematosus. Arthritis
Rheum 1982, 25:1271-1277.
15. Gladman DD, Ibanez D, Urowitz MB: Systemic lupus erythema-
tosus disease activity index 2000. J Rheumatol 2002,
29:288-291.
16. Hay EM, Bacon PA, Gordon C, Isenberg DA, Maddison P, Snaith
ML, Symmons DP, Viner N, Zoma A: The BILAG index: a reliable
and valid instrument for measuring clinical disease activity in
systemic lupus erythematosus. Q J Med 1993, 86:447-458.
17. Dennis G Jr, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC,
Lempicki RA: DAVID: database for annotation, visualization,
and integrated discovery. Genome Biol 2003, 4:P3.
18. EASE: Expression Analysis Systematic Explorer [http://
david.abcc.ncifcrf.gov/ease/ease.jsp]
19. Ingenuity Systems [http://www.ingenuity.com
]
20. Rovin BH, Wilmer WA, Danne M, Dickerson JA, Dixon CL, Lu L:
The mitogen-activated protein kinase p38 is necesssary for

29. Takeuchi O, Akira S: Toll-like receptors; their physiological role
and signal transduction system. Int Immunopharmacol 2001,
1:625-635.
Arthritis Research & Therapy Vol 11 No 1 Lee et al.
Page 10 of 10
(page number not for citation purposes)
30. Sabry A, Sheashaa H, El-Husseini A, Mahmoud K, Eldahshan KF,
George SK, Abdel-Khalek E, El-Shafey EM, Abo-Zenah H: Proin-
flammatory cytokines (TNF- and IL-6) in Egyptian patients
with SLE: its correlation with disease activity. Cytokine 2006,
35:148-153.
31. Funauchi M, Sugishima H, Minoda M, Horiuchi A: Serum level of
interferon-gamma in autoimmune diseases. Tohoku J Exp Med
1991, 164:259-267.
32. Siegal FP, Kadowaki N, Shodell M, Fitzgerald-Bocarsly PA, Shah
K, Ho S, Antonenko S, Liu YJ: The nature of the principal type 1
interferon-producing cells in human blood. Science 1999,
284:1835-1837.
33. Rood MJ, Velde EA Van Der, Ten Cate R, Breedveld FC, Huizinga
TW: Female sex hormones at the onset of systemic lupus ery-
thematosus affect survival. Br J Rheumatol 1998,
37:1008-1010.
34. Aringer M, Steiner G, Graninger WB, Hofler E, Steiner CW, Smo-
len JS: Effects of short-term infliximab therapy on autoantibod-
ies in systemic lupus erythematosus. Arthritis Rheum 2007,
56:274-279.
35. Palucka AK, Blanck JP, Bennett L, Pascual V, Banchereau J:
Cross-regulation of TNF and IFN- in autoimmune diseases.
Proc Natl Acad Sci USA 2005, 102:3372-3377.
36. Doria A, Iaccarino L, Sarzi-Puttini P, Ghirardello A, Zampieri S, Ari-